I would argue that component-level repair _is_ rapidly becoming a lost art. Apart from people on this list (and it's not even universal here) and people doing related hobbies like vintage radio repairs, component-level repair is alas very rare now. Even TV repairs (at least in the UK) have become board-swapping with the advent of the plasma panel and LCD TVs (most manufactuers of such sets do not supply schematics, and don't supply the ASICs other than on the PCB). What's more worrying to me is that the ability to diagnose a fault seems to have gone as well. We've had this rant to many times that I don't feel like starting it again... YEs, AFAIK there were never published schematics for the VAXstations. And since (from the few I've looked inside) they're full of custom silicon, it's going to be very hard to produce meaningful schematics of such machines. For a machine built drom known components (ICs you can get data sheets or at least pinouts for, etc), it's possible (although a lot of work [1]) to trace out the schematics. But II don't fancy doing it for a machine where the pinouts of the CPU and the main support ICs are unknown. [1] That is the voice of experience... -tony

Ture enough. There is a lot of differnce between design and repair (I've done both, and FWIW, I do not regard the former as being necessarily more difficult than the latter...). Of course there are some common skills for both -- you'd better know how the components behave, how to solder, etc. But equally, repair (or more precisely diagnosis) of a design that's worked once is IMHO very different from working out why your newly-created design doesn't work -- for one thing in the latter case I would hope you know how you inteded it to work in the former case you may well not fully understnad the circuit. ASICs bring 2 problems. THe obvious one is getting replacements, but the obvious answer to that is to raid them off old boards... The lkess obvious problem is that the behaviour of an ASIC will probably not be fully understood. Which means you may well not be able to tell which signals on the pins are correct, which are incorrect, and wheter the incorrect ones are caused by incorrect inputs or a fialure inside the ASIC. That makes fualtfinding a lot harder... If you take a VAXstatiuon board, you could produce a sort-of scheamtic. It would show the ASIC as a 200-pin (or whatever) device and show what each of the pins linked to. Some might be very understnadable (power, ground being the really simple ones). But then some might not, particularly if there are several ASICs with traces that just link them. Given that sort od scheamtic, you could, I suppose, find broken tracks (2 pins not connected that the schematic shows should be) or shorts (the reverse :-)). But other than that, it';s not a lot of help. If you stick a 'scope (or whataver) on a pin and see a particualr signal, you can't know if it's right or not, You can't know if the timing is reasonable. So how do you know what to replace? A _meaningful_ schematic (and in my experience a lot of commercial schematics are not 'meaninful' in this sense) firstly names all useful signals (clocks, address/data buses, bus strobes, etc) and secondly shoews circuits in the conventional way (e.g. a pair of cross-coupled NAND gates froming an SR flip-flop is shown just like that, you don't end up with the 2 gates on different sheets). Having done it a few times, I am convinced that producing a meaningful schematic is a lot of work... -tony

I've acquired a small cache ( 100+ ) of the IMSAI style C&K P/N 7101 paddle style switches ( with Black paddles ). Digikey wants $7.70 and other places want even more. I can let them go for $5.00 a piece, maybe less if you want a whole bunch. Just wanted to give the group a shot at these before I throw them to the great un-washed on Epay ..... Please contact me offline from here so we don't piss anybody off. Best regards, Steven

On Mon, 25 Feb 2008, Roy J. Tellason wrote: Too bad my plan to make a CPU from TTL isn't anywhere near the stage to go sourcing parts. -- David Griffith dgriffi at cs.csubak.edu A: Because it fouls the order in which people normally read text. Q: Why is top-posting such a bad thing? A: Top-posting. Q: What is the most annoying thing in e-mail?

Grant Stockly wrote: Somebody send him a PDP/8e to scan ... We could use another vintage replica. :)

Apologies if this appears twice, but it looks like hotmail ate my message again. From: "Grant Stockly" <grant at stockly.com> I've got modern CAD drawings for a replica backplane for the 8/S, if that's of interest. That and about a dozen module drawings will let you build a replica (given enough cash). Of course, to be authentic, you'd export the netlist and wire wrap the backplane, but that's a preposterous amount of error-prone work. I do have a Perl script to generate a wirelist, though, if someone wants to try it :-). Even though the modules are all single-sided, that's a lot of wire-wrapping! The replica also uses an SRAM memory module, instead of the original core. The SRAM module is double height (to get enough pins), and standard DEC modules are used to interface to it. I haven't done anything with the slots freed up by replacing the core subsystem. (There might be enough room to integrate a TTY controller!) I do have the TTY controller backplane drawn, too. Oh, and the memory expansion above 4K is also still a separate backplane (which I haven't drawn). I do know of a guy in Australia who's done some work on replicating the Rxxx modules. Generally, I think the 8/L backplane is a more promising starting point, though it lacks the cachet of being all-transistor (and the DEC modules use dated, hard-to-find TTL). Of course, the memory expansion for that is *also* another backplane (which I haven't drawn). Vince

From: "Roy J. Tellason": Cool! Depending on what you are working on in the Mxxx series flip-chip line, there's a bunch of 74Hxx, quite a bit of use of and-or-invert (7450, 7453, H50, H52, H53, H55) and expander gates (7460, H60, H62), 7482 adder, 74181 alu, 74182 look-ahead carry. If you are specifically building an 8/L, you can get away with a subset, but I know off the top of my head that the M220 card (major registers and ALU functions) uses 7453, 7460, 7482, and I believe 74H40 chips (as well as the 7474, which isn't too hard to find). Of course, you could redesign the flipchips to use more modern stuff, but then you might get into a semantic quibble about whether it was really still a PDP-8/L :-). And you probably still need a fair number of 7401's or PALs or whatever to replace the and-or-invert gates and arithmetic stuff. I have some of these components, but the H5x and H6x stuff I mentioned I've had trouble sourcing even a couple of. You can also often get stuff for prices like $5 per chip, but that wouldn't be really practical if you were trying to build an entire replica CPU out of them. Vince

David Griffith wrote: Also don't the have programable interconnects now days too- forget wire wrap. Oh wait we now also need tiny DEC-TAPES too. :)

On Feb 26, 2008, at 11:33 PM, David Griffith wrote: I don't know who else is thinking about it, but I hope to do some work in that area eventually. -Dave -- Dave McGuire Port Charlotte, FL

Vincent Slyngstad wrote: I'd be interested in a system that used the CF connectors to be the edge card connectors. they should be cheap, and would hit the form factor well. Both sexes are available so you could integrate the backplane with the connectors prewired on a big pc board, and make gobs of the little flip chips to plug in. I'd be interested in a tiny front panel with SMT LED's but don't know what you'd get for the front panel switches. Jim

On Wed, 27 Feb 2008, jim s wrote: Another option is PCIExpress 1X connectors, They are cheap and more rugged than CF sockets, and 36 pins! Peter Wallace

On Wed, 27 Feb 2008, Vincent Slyngstad wrote: Probably cheaper/board than the female part of CF socket... Even with gold, 2 or 4 layer PCBs this size should be less than $2.00 each Peter Wallace

Hmm. I get $5 per board, and that's assuming you can get quantity 100 of each of the module types! For a single replica, you'd pay more per board. I didn't get prices around $2 until quantities closer to 250 per design. Unless you know some place a lot cheaper than I do. From: "David Griffith": That part's fairly easy. As I mentioned, I've got a DEC style backplane already drawn for a variation on the 8/S. All that's really required is to define a new package and "change package" on all the components. One issue that comes up with the DEC backplanes is that all the connectors are right next to each other, leading to pretty dense wiring on the backplane. Most DEC backplanes I've tried to route as PCB's take at least 4 layers. They're also really big boards -- typically 12"x16" or more. Which means they're a few hundred dollars each in small quantities. If these are half-size (or less), maybe they'd drop back into the size range where the inexpensive fabs will do them. (Most of those seem to stop at 10"x12" or so.) We *do* all understand that an 8/S replica is a multi-thousand dollar proposition, right? If the boards could be done for $5 each, and the connector is another $0.50, and the components another $2 per board, then we are looking at about $7.50*400=$3000 already, and we haven't got a backplane or a front panel, each of which is probably a few hundred dollars. Another hundred for the power supply, and gotta buy a pretty box. Even with prototype board pricing and cheap interconnect I never came up with cost estimates below about $3000 for a whole system (which is partly why I never tried to build any of it). Quantity discounts would make a big difference, but even so I don't see it getting below the $2000 mark. It'd be great to be proved wrong :-). Vince

From: "Peter C. Wallace": I don't know any place that will let you prepanelize and also do gold fingers. Which fab is that? Prepanelization would make the module PCBs *much* more affordable. Vince

From: "Peter C. Wallace": I couldn't find how to get quotes from them, so I guess I'll have to take your word for what you can get the PCB's for, then :-). Vince

From: "Grant Stockly": Yeah, but you can't ask for gold fingers at that price. Vince

Peter C. Wallace wrote: excellent suggestion. My suggestion needed an additional connector, and would offer nothing by having it. to another point brought up, I was only worried about enough fingers, not too few. I have access to a couple of Standard wire wrap machines, and the lady who used to run it, if anyone can get an idea how to do the boards, and get the paper tapes. I wonder if it is worth the effort of wirewrap, since the connectors don't have wirewrap tails? It's been a while since I looked at what goes on with the DEC backplanes and wirewrap. Do they populate pins regardless of the connectors having wirewrap tails, or are the pins mostly from the connectors? We had a system which used a lot of wirewrap boards, (off these machines) but we never lacked for PC backplanes and card cages due to a good surplus buy. So I didn't have to wirewrap any backplanes, thank goodness. Jim

Vincent Slyngstad wrote: Well a real straight 8 to me sounds better than a S.

On Feb 27, 2008, at 4:33 PM, jim s wrote: Even two-row 0.1" header pins would probably work well. That's what I'll most likely use for a backplane-based Z80 system I've been working on for a while. "Real" backplane connectors are very, very expensive. -Dave -- Dave McGuire Port Charlotte, FL

On Feb 29, 2008, at 3:37 AM, Brent Hilpert wrote: No, not in this case...I'm designing what one could call a "hybrid" of old and new bus techniques. It carries the standard Z80 address, data, and control buses, but it has such things as out-of- band board identification capability. The system will have a "motherboard" on which the CPU, memory, and console ports reside, and the bus slots will be for expansion cards. Wow, that sounds like a fun machine! Why don't I ever find cool stuff like that anymore?! -Dave -- Dave McGuire Port Charlotte, FL

From: "Jos Dreesen / Marian Capel": Well, there's FRAM, which I guess would keep yucky batteries out of the picture. But I don't have any ideas about where to get real core planes these days (except in quantity one). Vince

On Thursday 28 February 2008 15:35, Sridhar Ayengar wrote: Dunno, but in the past day or two core came up in my list, and "that auction place", so I looked over there and somebody's actually selling containers of just cores. Not a job I'd care to tackle, for sure! -- Member of the toughest, meanest, deadliest, most unrelenting -- and ablest -- form of life in this section of space, ?a critter that can be killed but can't be tamed. ?--Robert A. Heinlein, "The Puppet Masters" - Information is more dangerous than cannon to a society ruled by lies. --James M Dakin

From: "Chuck Guzis": Agreed, though iti is intrinsic in all the transistor and TTL PDP-8 designs to always write after read. So you'd have to either let the writes through, or suppress them within the replacement memory subsystem itself. Vince

Christian Corti wrote: Hoo boy, you reminded me of The Jones Machine, by one Hillary Jones (no relation). Published in Byte in the 1980s, a 4 instruction / 8 bit machine using a 74181 and 7489s. IIRC, one instruction was "store the 1's complement of the register" which was more or less a freebie from how the 7489s were used. A project I did for college was expanding it to 16 bit words, four bit opcode and twelve bits of address, using more modern/common TTL (circa 1987), and adding instructions to support a stack, calling subroutines, and ...? Boy, I wonder where the paperwork for that is. Never got around to trying to build one so it's just a paper design, but something I was rather proud of at the time. --S.

New information about the workings of the Antikythera mechanism. I do believe it counts as vintage enough. LOL http://www.guardian.co.uk/science/2006/nov/30/uknews Greg

Come now. The differential gear train that subtracts the revolutions of the sun, and correctly computes the synodic month of 29 1/2 AND exactly duplicates the 19 year pattern of equinox and moon phases counts as computing. Read up on Derek De Solla Price. Randy _________________________________________________________________ Connect and share in new ways with Windows Live. http://www.windowslive.com/share.html?ocid=TXT_TAGHM_Wave2_sharelife_012008

OOPS! Better check my dates better next time. Sorry. LOL Greg

To answer a few questions that have been asked, this part ( 7101 ) is a SPDT non-momentary switch used in the bus toggle position. The control switches were the momentary ( P/N 7105 ) of which I unfortunately have none. Hope this helps and thanks for the questions. Personally, I like the Black ones better than the red and blue, but that's just my preference. Best regards, Steven paddle places want throw from

On Monday 25 February 2008 02:08, Alexandre Souza wrote: The continuing ability to make a living at something I'd spent several decades acquiring and honing my skills for would be nice, too... -- Member of the toughest, meanest, deadliest, most unrelenting -- and ablest -- form of life in this section of space, ?a critter that can be killed but can't be tamed. ?--Robert A. Heinlein, "The Puppet Masters" - Information is more dangerous than cannon to a society ruled by lies. --James M Dakin

On Tuesday 26 February 2008 06:18, Alexandre Souza wrote: Ah, but I do know tubes, transistors, chips, logic-based stuff, and even microprocessor/microcontroller-based stuff as well. The problem is that the amount of work needed out there in the market has gone way downward with each step in that progression, with less and less of it to be found as time goes on. Most of what used to be around here for repair shops and such don't exist any more. That last part is the issue, sometimes. I do keep on learning new stuff all the time, and I look for situations to apply it in, but the job market at least as far as I've been able to find anywhere around here with regard to someone who's self-taught flat out stinks! But I'll keep on looking... -- Member of the toughest, meanest, deadliest, most unrelenting -- and ablest -- form of life in this section of space, ?a critter that can be killed but can't be tamed. ?--Robert A. Heinlein, "The Puppet Masters" - Information is more dangerous than cannon to a society ruled by lies. --James M Dakin

On Mon, 25 Feb 2008, Tony Duell wrote: We have the original HP blueprints/schematics for the HP 9810 and/or 9820. I haven't scanned them yet as they are on big sheets. I should try to copy/reduce them onto A4 and then scan them. Christian

Tony Duell wrote: Maybe, but I think that bitsavers has schematics for at least one of the MicroVAX 3100 series machines. These are of the same vintage as at least some of the machines in question. For the Alpha-based machines, there's plenty of information. I think that the EV4 and EV5 pinouts are floating around. I would be surprised if the EV6 stuff were not too. Some of the VAX CPUs and support chips did appear in various Semiconductor Volumes, some of which are also floating around. It would still be difficult to fix a fault in such a chip (even if a replacement could be found) since they at some stage went to BGA and even access to the appropriate tools + experience doen't get you a success rate much above 50%. That being said, I didn?t see sufficient information or detail in the posts so far to suggest that _anything_ is actually broken as opposed to being misconfigured or misused. I think the OP would do well to state exactly what is believed to be working and then pick on one of his apparently broken machines. It may be that some simple reconfiguring is all that is required. Or maybe the machine can be stripped to the bare minimum to see if it comes up then (I think I've mentioned before the mouse on my VS3100-76 that stopped it powering up at all ...). Antonio No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.5.516 / Virus Database: 269.20.9/1295 - Release Date: 23/02/2008 21:35

Well, from a Montreal perspective, there are three alternatives I'm familiar with: Addison, a surplus discount house; Active, which last time I went there was a parts house like what RS was back when only better; and order-in, like Digikey or Mouser. Oh, and for really common routine things like connectors and switches, places like The Source often have a parts corner - though I fear it's headed the way of Radio Shack's parts section. As for who knows how any longer, the first thing that comes to mind is rock band guitar techs, probably because one of the few electronics techs I know who worked in a job which still exists was one. /~\ The ASCII der Mouse \ / Ribbon Campaign X Against HTML mouse at rodents.montreal.qc.ca / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

On Sunday 24 February 2008 10:32, Dan Gahlinger wrote: There's a few different methods. Which did yo use? You don't need to have a valid license to log in as "system" on the console. Pat -- Purdue University ITAP/RCAC --- http://www.rcac.purdue.edu/ The Computer Refuge --- http://computer-refuge.org

At 08:47 PM 2/24/2008, Dan Gahlinger wrote: No, actually, you don't. Console login doesn't require a valid license. See the OpenVMS FAQ. Please note that "console" does not mean the graphics device. You will get a "nolicense" error if that's the case. Perhaps if you use the Open *VMS* FAQ it'd work better. :-) After "spawn", you forgot "@SYS$SYSTEM:STARTUP". Without that, the necessary protected shared libraries won't be installed. However, you probably did execute that, as you can't run authorize without those being installed. You may have a disabled system account or some other problem. Perhaps just using UAFALTERNATE set to 1 will work. In AUTHORIZE, "mod system/password=whatever/nopwdexpired/flags=(nodisuser,nocaptive,nopwd_expired,norestricted)" Those are the most likely bogon flags that could be causing you problems that I can think of. -Rick

Dan Gahlinger wrote: Not as SYSTEM on OPA0:. Quite what counts as OPA0: on workstation depends on how your console switch is set. There are, as always, gotchas and quid-pro-quos. The FAQ tells you to: OK. No. It's more like: $ SPAWN $ @SYS$SYSTEM:STARTUP Then, when that's done, you $ SET DEFAULT SYS$SYSTEM If you don't do that bit you probably end up creating a new UAF file in whatever directory you left yourself in. Then you'll quite happily go and modify that, but the system won't use it! Now you can run authorize and fiddle. But even there you need to do a SHOW SYSTEM afterwards and make sure that it hasn't been DISUSER'd or whatever through too many failed attempts or some other bad flag hasn't been set. It's also possible (although generally unlikely) that the system uses an alternate UAF. Have another go .. at least this system works! Antonio No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.5.516 / Virus Database: 269.21.0/1296 - Release Date: 24/02/2008 12:19